ZA | 25-SDC-July | Luke Manyamazi | Sprint 2 | Shell Pipelines Exercises

[Luke-Manyamazi]

Learners, PR Template

Self checklist

• I have committed my files one by one, on purpose, and for a reason
• I have titled my PR with Region | Cohort | FirstName LastName | Sprint | Assignment Title
• I have tested my changes
• My changes follow the style guide
• My changes meet the requirements of this task

Changelist

Practiced listing and filtering files using ls and grep, including matching patterns and using extended regular expressions
Used sort, uniq, head, and tail to process text files like scores-table.txt and events-with-timestamps.txt:

• Sorted lines by specific fields (e.g., person names).
• Counted unique entries and exits using a combination of tools.
• Extracted top/bottom lines of sorted data for analysis.
  Explored tr to transform and clean input data streams, such as converting text to uppercase and replacing whitespace.

Questions

1. Is my usage of sort -k to sort by names correct and efficient, especially for larger datasets?
2. Are there more concise ways to combine uniq and awk to get counts for multiple patterns (like Entry/Exit) more elegantly?

[LonMcGregor]

Hi, good work. I have one comment in task 4 of ls-grep for you to review. Also, can you try to complete the sort-uniq tasks without using grep?

[LonMcGregor]

Do you know of any arguments for grep that would let you count without needing to pipe into wc?

[Luke-Manyamazi]

From the research I did, it seems that non-standard versions of grep, like GNU grep, support the -c option, which counts the number of matching lines.

I could use a regex together with grep -c as an argument, as below:

ls sample-files | grep -c -E '^[A-Z]$|^[A-Z][a-z]*$'

[Luke-Manyamazi]

Thank you, @LonMcGregor, for reviewing my exercises. I really appreciate your feedback and guidance.